Did you have to actually play any of them?

If you have a sufficiently-large bankroll (which you probably do), and you have developed a DoN strategy which takes advantage of a high-variance playing style within the context of DoN's to increase your $EV, then there's nothing wrong with this.

On the other hand, it might be possible to refine your playing strategy to maintain your same or even a higher $EV while reducing the variance experienced.

After 2k game I only have 4.5 % ROI, but that includes a downswing of 47 buy-ins and one of 34 buy-ins, so hopefully my true ROI is a few points higher and it should at least be 6 % so don't be scared, just be prepared for the worst!

re-read the thread plz.

Higher EV = lower variance in DoN.

Not that your understanding of variance seems that great.

How do you know what it "should" be?

When you played tens of tousands of SnG's you understand that there's a big spread between good and bad 2k runs and even though it's hard to know how bad your latest run has been becuase you may have been really lucky in between your downswings you often can get a pretty good read. I'm pretty sure my run is below my true ROI. The swings I been through over these 2k games are so sick compared to most people playing DoN's that overall my run should be on the lower half and I would estimate that I will have runs over 2k games with a ROI above 10 % as well, but of course it's hard to say how much higher my true ROI is, but if I had any money I would bet pretty hard on >7 % after 6k games :-)

Ever notice how this thread swings back to variance every two days or so?

What surprises me is the number of people who skip the thread entirely and just post. Yeah, there are almost a thousand posts in this thread but if I'd just started playing DoNs I'd have absolutely (obsessively) read every post in this thread. Particularly if I was about to post something that's almost certainly been covered dozens of times before. Not just so that I don't look like a complete n00b but because there's probably a ton of good information in a thread this size.

So, if you're one of those guys who didn't bother reading this thread and instead just dove in posting a blindingly obvious question - Maybe you should just forget about it, you don't have the kind of drive and/or interest to really excel at this.

Merry Christmas everyone!


Yes, everyone who's playing high volume DoNs has experienced big and/or protracted downswings. While DoNs are less variance per earn than other forms of poker that cuts both ways. Massive rushes are not as common either and it's more difficult to recover from a downswing. In addition, when you do wind up playing for stacks, you're often only a slight favorite (60:40) or flipping. That can lead to some very brutal runs. Remember, DoNs are really about consistency. You're trying to win only slightly more tournaments than you lose. You can't have one day where you play amazing and then the next day play your B game if you want to do well at DoNs.

hands of the day:


Poker Stars $5.00+$0.20 No Limit Hold'em Tournament - t50/t100 Blinds + t10 - 9 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com

SB: t2760
BB: t1510
UTG: t1810
UTG+1: t1355
UTG+2: t1485
MP1: t1500
MP2: t1470
CO: t1795
Hero (BTN): t1315

Pre Flop: (t240) Hero is BTN with T A
6 folds, Hero...

------------------------------------------------------------


Poker Stars $5.00+$0.20 No Limit Hold'em Tournament - t100/t200 Blinds + t20 - 8 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com

BTN: t470
SB: t5855
Hero (BB): t1385
UTG: t900
UTG+1: t3020
MP1: t1085
MP2: t1535
CO: t750

Pre Flop: (t460) Hero is BB with 9 K
UTG raises to t880 all in, 6 folds, Hero...

I got to a flop BvB and immediately c/f ... ended with over 1400 in chips without ever winning a hand.

I completely agree. Although I'm not one of the higher volume guys here I do have about 2500 of these things done so far and am putting them away at 100+ a day now (stacked sets ftw). Definitely see big differences in A game versus A- game results ... although table selecting isn't really possible, avoiding other known good regs (emphasis on good) by only starting new tourneys and letting them fill helps some too.

lol @ not reading through a ridiculously sized thread says much at all about your ability to succeed at much anything in life.

Also, CMAR does not seem to really grasp how these pokers work out. The factoid that we have smaller edges when we end up AI (I think this is what he's saying, and I obviously think it's just as ridiculous as the next guy thinks) in DoNs compared to 50/30/20s has got nothing to do with our variance.

Why do I have to post this on every damn page of this thread?

Our variance in DoNs is only related to our ITM%. Our ITM% is only related to our skill (and well, our variance, if you want to go down that route). Every decision that is higher variance in a DoN is also by definition -EV.

Consider that paragraph until you understand it. Fully. There's talking about reading through 10 pages. LOL.

Also, I'm curious to hear an explaination as to why it's particularly bad in DoNs to play poorly one day.

If anything it can't matter much, as everyone who plays them is so bleeding bad.

As much as I think you're being a prick, I have to agree.

Especially, I can't think of a single reason why playing below your best is worse in DoNs- in fact I think it's the opposite. The skill differences between fish, average regs, and solid regs/pros in these are absolutely staggering from a strategy standpoint, yet everyone experiences relatively low variance.

And to keep this a strategy post, as a public service announcement, please understand that it's ok to fold AK. I'm seeing AK and AQ being played so fuggin atrociously by everyone from robot regs to clueless LAGs lately. If you have 2700 chips with seven players left and two shortys and you're in 3rd place, why in God's ganja garden are you calling every T1900 shove? I won't even go into the merits of sometimes ditching big slick to a single standard raise.

I don't understand the last sentence. Higher variance than what?

You just said variance is related to skill. No it isn't. How much variance you can expect is related to skill I suppose. Variance good or bad is just the short term correcting for the long term.

Lets say we get it all-in as a 60:40 100 times. We win exactly 60 and lose exactly 40 - but we might win the first 60 - and then lose the next 40 - we have run exactly at expected value but our first 60 flips were awesome and the following 40 were "brutal".

Variance is how much our expected results vary in the short term from the long term mean expected value.

If you get it in a lot as 55/45 60/40 it is much more likely that you will get a run of all losses in any give sequence of flips.

e.g. consider a roulette wheel - the red/black bets at ~50:50 have a muchmuch higher chance of repeating for a long sequence of reds than the 1-12/13-24/25-36 bets at ~1:3 have of repeating a long run of 1-12's

Your skill defines how good you get your chips in on average when it matters, if on average you are a 55/45 favourite in those spots you will experience more variance than if you are a 59/41 favourite on average, thats about it. What variance is has no relation to skill - how much you can expect to encounter does.

When I am playing my A+ game I read peoples souls and make well timed calls and shoves and am usually no worse than a 70/30 favourite in critical spots. - When I am playing my B game I am often only a 55/45 favourite - when the edges we are playing with matter a lot that makes a hell of a difference to my bottom line - YMMV

and to offset yet more variance discussion here's some strat:

Early doors - blinds 10/20 I have t1500 as does villain.

Villain opens to t60 from MP I flat call OTB with 44

Flop comes monotone K 8 4 pot is t150 (2players)

villain checks , I bet t125, villain calls.

Turn 9

check, check

River 3 pot is t400

Villain snap pots it and I ..... ??


I used to shove this river - or bet enough on the turn to setup a river shove and be very happy to gets it all in. After all thats exaclty how I would play a set at a cash table.

I have started playing a lot tghter and slower at early blind levels with good but nonnut hands, with the rationale that If I lose a small pot and drop my stack to t1000 or so very often I am able to recover. However most of the time it seem that villains who lose a 1/3rd of their stack early tend to be one of the first out.

Is ths a bad adjustment I have made ? and should we be playing early levels like a cash game, or is it better to tighten down and preserve the stack as much as possible until it gets to the push/fold stage...

On the contrary, I think you've just discovered that there is no spoon ; )

I've got a query for good players regarding a general situation. In BvB situations, late when I am sitting on a tenuous stack that still has FE but only for one full round, the SB will so often limp/minbet flop into my BB that it feels like rape. If I hit top pair I'm usually shoving, obviously, but how about middle pair/bottom w an over or a probable 40/60 draw? We can call and see if they check turn, but obviously that is so passive we're only using that against very specific opponents and not with a weak draw. I have been randomizing here when our stack and reads do not dictate a certain action, something like 75% fold and 25% shove, leaning toward the fold with exploitable stacks to the left and leaning toward the shove with big stack callbots on my left. Any input you might have is greatly appreciated- I am all about giving myself a $0.25/hr raise.

No need to be snippy.

"Variance" and "EV" (expected value/expectation) are mathematically independent. In general, however, there's a trade-off between higher variance in returns and expected value, but it's in the opposite direction that you think it is. "Investors" (e.g. poker players) who are subject to a higher level of variance to obtain their expected rate of return require that rate of return to be higher to compensate for the greater volatility of the expected return.


If the variance in DoN's is lower, than the expected returns must also be lower, at least in comparison to other forms of poker with higher variance.

And indeed, the expected return in a DoN for the winner is, at most, regardless of skill level, 1/5 of all the $$$ equity available. That's lower variance. But's it's also a lower expected return.

Q.E.D.

"Variance" is a mathematical/statistical concept which helps describe the expected or hypothetical distribution of expected returns over the distribution of the returns of all players. Attributing an individual downswing of an individual player (e.g. a particular "bad run") to mathematical "variance" is fallacious.

Mathematical variance is that portion of the player's result which can be attributed to totally random events over which the player has no control (i.e. the "fall of the cards.")

Let's say Player A has a 40 bi downswing. Some of that is due to mathematical variance, undoubtedly; but how much? How do we know the player just wasn't "playing bad" for a while?

Yes I have read many (but certainly not all) of the posts on this particular thread. I don't have that kind of time. But certainly many of the people who have posted don't seem to have even a vague understanding of what "variance" actually means. So we even have very experienced, highly successful players who seem to believe that lower variance = higher returns.

Perhaps if they had a better understanding of what variance actually means, they would be able to adjust their strategies accordingly.

Awesome! Vng.

The average starting position for all 10 players in a DoN is 5.5, i.e., halfway between 1st place and 10th place (1 + 10 = 11 divided by 2).

In order to end up "in the money," the player has to "move up" half a place, i.e. to 5th place.

So, on average, how many chips are needed to end up in 5th place?

Does anyone know the answer to this empirically? What's the average number of chips that the 5th place-holder has at the end of the average tournament?

Or, looking at it from another angle, how many DoN's actually end when the 6th place finisher simply runs out of chips due to blinding and/or anteing out? As opposed to making a "bad decision" to put the player's last remaining chips into the pot (of necessity an "all in" on a bet/fold/raise)?

Clearly, in a DoN, the player can never lose if the player never goes all in. Even when blinded or anted out, the player is forced to bet all in.

This suggests that the player should be extremely cautious about going all in voluntarily, as the only way one can ever lose is either by going voluntarily all in, or forced to do so by antes/blinds.

It seems to me an optimal strategy might be to try to always end up in fourth place chip-wise. Or maybe third place at most. Fifth place being a little bit too close to the bubble, no margin for error. On the other hand shooting for anything higher than third place is simply taking unnecessary chances for chips that don't add to $EV.

Clearly there doesn't really seem to be much of a need to ever get more than t3000. After that you could probably fold every single hand and not blind out before the bottom 5 have been knocked out. But perhaps I'm wrong about this, I've just started looking into this.

So assuming once you've attained t3000 you're pretty "safe," the question then becomes, how much less than t3000 does the player need to acquire, on average, so that they could then theoretically fold every hand and not blind/ante out prior to winning?

This obviously depends upon the length of game. Someone posted earlier about a short game in which they never played except to c/f one hand.

Making some average assumptions (45 minute DoN with the 'normal' times for hand length and players dropping out) I came up with the blinds and antes equalling about 3250 for the game as a first approximation. This is an artifically low number though for at least two reasons. First, this assumes 'about' one orbit per blind level which is sometimes not the case in foldfests and secondly, if you were to double up plus in the first hand to say 3500 and try and fold your way through, you'll be the shorty or near shorty in the mid to late levels -- this will change the behavior of the other parties. I know I've folded KK, AK with 2BB to my name because someone else was in harm's way already (I also know I did NOT fold the AA when I should have but I'll keep the bad beats out of it).

As with most things poker related -- it all depends.

Malthus, I believe trying to figure out how many chips you need is of minimum value, and might tend to force a player into achieving a certain goal rather than playing each situation optimally.

I am massively -c over thousands of DoNs and I find this to work out fine. I rarely finish over T3000 and am often in 5th on the bubble and play very comfortably from there.

O.K.

This suggests that once itm there's really not much point in risking chips in hand for more chips, I guess other than to keep up with the blinds/antes.

So when you get to t3000+ you're pretty much folding everything?

my thoughts on swings categorized by the skill of player "champion don player" average downswing for 20.80 12 game downswings ,for 52s 16 game downswings,104s 25 game downswings."very good player with little weaknesses" 20.80 15 game down swings,52s 20 game downswings,104s 30 games swings ."winning players that make tons of - ev calls " 20.80 20 game swings,52s 25 game downswings,104s 35 game downswings .'' break even players" 20.80 25 game swings ,52s 30 games swings,104s 40 game downswings .Everyone wants to post about how bad their swings are and they think their having really crappy luck when they go on a 40 game down swing in the 10s and 20s,the truth of the matter is if you have a 40 game downswing in the 10s 20s,your making tons and tons of mistakes.Of course good champion players who make almost no mistakes are going to have swings,they will just have way fewer swings than all of the other caterogized players above . Minimal roi i think the best don players should have in the 20s would be 10 % ,8% in the 52s and 3-4 % in the 100s.Anything less than this and you are not playing the best mathematical winning strategy

This seems to me to be pretty read-dependent. There is no magic formula.

Sometimes you get down to 7 players and you will see 5 or more consecutive all-ins go unchallenged. Other times, there are never 3 all-ins in a row without a call.

One of the keys is the big stack. If someone with 4K turns down 2 chances to call all-ins by short stack, it's a lot safer to shove when you have a hand.

By the way, my successful DONs have ranged from under 20 minutes to over 90, so generalizing about how many chips you need to make it through is rather difficult. I remember someone posting that he made it to blinds of 300/600.

These things can be as much an art as they are a science.